PRICAI'98: Topics in Artificial Intelligence 5th Pacific Rim International Conference on Artificial Intelligence, Singapore, November 22-27, 1998, Proceedings

This book constitutes the refereed proceedings of the 5th Pacific Rim International Conference on Artificial Intelligence, PRICAI'98, held in Singapore, in November 1998.
The 56 revised papers presented were carefully reviewed and selected from a total of 197 submissions received from 25 countries. The papers are organized in sections on induction; multi-agent architecture; knowledge acquisition, modeling and validation; reasoning; knowledge discovery and data mining; knowledge management; application of fuzzy logic; applications of neural networks; searching; Bayesian networks; text analysis; knowledge structure; inductive logic programming; intelligent systems; foundations of AI; stochastic processes; abduction; information retrieval and speech recognition.

Repechage bootstrap aggregating for misclassification cost reduction.- Generating classifier committees by stochastically selecting both attributes and training examples.- Multi-layer incremental induction.- An adaptive agent oriented software architecture.- An architecture for multi-agent negotiation using private preferences in a meeting scheduler.- Towards agent-oriented smart office based on concurrent logic languages.- Analyzing the roles of problem solving and learning in organizational-learning oriented classifier system.- Simultaneous modelling and knowledge acquisition using NRDR.- Model building and program specification in a case of enterprise design.- On the practicality of viewpoint-based requirements engineering.- Reasoning without minimality.- Reasoning with sets of preferences in default logic.- Frugality in reasoning and the role of summary.- Data mining for risk analysis and targeted marketing.- Applying knowledge discovery to predict infectious disease epidemics.- Using decision tree induction for discovering holes in data.- DODDLE: A domain ontology rapid development environment.- CAMLET: A platform for automatic composition of inductive learning systems using ontologies.- Management of worker’s experiences: A knowledge-based approach.- The minimum description length based decision tree pruning.- Hybrid search of feature subsets.- Concurrent discretization of multiple attributes.- The integration of machine and human knowledge by fuzzy logic for the prediction of stock price index.- Computational intelligence techniques for short term generation scheduling in a hybrid energy system.- Fuzzy rules extraction based-integration of linguistic and numerical information for hybrid intelligent systems.- Reasoning about continuous change.- Point-based approaches to qualitative temporal reasoning.- An experimental study of reasoning with sequences of point events.- A mixture of global and local gated experts for the prediction of high frequency foreign exchange rates.- A neural network diagnosis model without disorder independence assumption.- Neural network based motion control and applications to non-holonomic mobile manipulators.- A geometric approach to anytime constraint solving for TCSPs.- Dynamic constraint weighting for over-constrained problems.- Time-constrained heuristic search for practical route finding.- Using mutual information to determine relevance in Bayesian networks.- Context-specific independence, decomposition of conditional probabilities, and inference in Bayesian networks.- Derivational grammar approach to morphological analysis of Japanese sentences.- An automatic Thai lexical acquisition from text.- Knowledge representation issues in information extraction.- Structuralization of case-bases, using concept hierarchy.- A closer look at preduction.- Learning linearly-moded programs from entailment.- Learning first-order rules from image applied to glaucoma diagnosis.- Knowledge-based formulation of dynamic decision models.- An intelligent job counseling system.- The layout problem: Investigation and aggregation of artificial intelligence and optimization techniques.- Logical fiction: Real vs. ideal.- Coherence measure based on average use of formulas.- A decision-theoretic approach for pre-sending information on the WWW.- Dynamic non-uniform abstractions for approximate planning in large structured stochastic domains.- A characterization of contrastive explanations computation.- SL method for computing a near-optimal solution using linear and non-linear programming in cost-based hypothetical reasoning.- Image retrieval system using KANSEI features.- A robust front-end for telephone speech recognition.